PROJECT ABSTRACT Above-knee amputees often struggle to perform the varying activities of daily life with conventional prostheses. Emerging powered knee-ankle prostheses have motors that can restore normative biomechanics, but these devices are limited to a small set of pre-defined activities that must be tuned to the user by technical experts over several hours. The overall goal of this project is to model and control human locomotion over continuously varying tasks for the design of agile, powered prostheses that require little to no tuning. The universal use of different task-specific controllers in current powered legs is a direct consequence of the prevailing paradigm for viewing human locomotion as a discrete set of activities. There is a fundamental gap in knowledge about how to analyze, model, and control continuously varying locomotion, which greatly limits the adaptability and agility of powered prostheses. The central hypothesis of this project is that continuously varying activities can be represented by a single mathematical model based on measureable physical quantities called task variables. The proposed project will be scientifically significant to understanding how humans continuously adapt to varying activities and environments, technologically significant to the design of agile, user-synchronized powered prosthetic legs, and clinically significant to the adoption of powered knee-ankle prostheses for improved community ambulation. The proposed model of human locomotion will enable new prosthetic strategies for controlling and adapting to the environment, which aligns with the missions of the NICHD/NCMRR Devices and Technology Development program area and the NIBIB Mathematical Modeling, Simulation, and Analysis program. The innovation of this work is encompassed in 1) a continuous paradigm for variable locomotor activities that challenges the existing discrete paradigm, 2) a unified task control methodology that drastically improves the agility of powered prosthetic legs, and 3) a partially automated tuning process that significantly reduces the time and technical expertise required to configure powered knee- ankle prostheses. This continuous task paradigm will provide new methods and models for studying human locomotion across tasks and task transitions. This innovation will address a key roadblock in control technology that currently restricts powered legs to a small set of activities that do not generalize well across users. The adaptability of the proposed control paradigm across users and activities will transform the prosthetics field with a new generation of ?plug-and-play? powered legs for community ambulation.